Simple Resistance nightmare.

[kxenos]

Put in series a 2,2?, 1W resistor and it will be fine

[iamnothim]

I realize the heater isn't too picky about the voltage and the current.  The reason I'm trying to be accurate is I want to test a valve and compare it to the factory specifications. (see attached example)
In my small amount of reading, the heater heats the cathode resulting in electrons being emitted.   My reasoning is that variations in the power dissipated by the heater might have a result in the cathode giving up electrons.
That's why.   Might be crazy, but I'm trying to have an instrument in the end.   That's the fantasy.

All but the PCC88 are 6.3V but the current varies between .300A and .600  so I'd like to "dial it in" based on the valve I am testing.

Here's the bridge rectifier I'm looking at.  All the others had much higher voltages.
Purely a guess...

[David_AVD]

Maybe something with a higher current rating considering the inrush (surge) current when the valve is cold.  4 Amp bridges are still cheap.

[IanB]

..... Not getting PTC.....  I'm missing way too much.

Here is a plot that I just made for a 12 V bulb. I was interested in what numbers I would discover.



Looking at this, the first data point is 0.1052 V, 0.107 A giving effectively 0.983 ohms. The last data point is 14.0 V, 1.076 A giving effectively 13.01 ohms. So the effective resistance increased by a factor of about 13x from cold filament to hot filament.

Another interesting number is the incremental or localized resistance between 13 and 14 V. The numbers are 13.0 V and 1.035 A, 14.0 V and 1.076 A. So the local resistance at this point is about (14 - 13) / (1.076 - 1.035) = 24.4 ohms. We could say that the "stiffness", or regulation of the bulb at this operating point is 41 mA/V.

[iamnothim]

Ian,  Outstanding! 
I love graphs.  This one illustrates what Richard was writing about and what I had a glimpse of with my max/min.
It's very interesting.

Through the first volt you really see the surge in current.

Since I caught the 1.5A burst quickly, I'd love to see it over time.
I'm playing the idiot card here.   I've been at this 8 weeks tops... probably only tested continuity and voltage but not across anything...

This would be a worthwhile diversion for me.... Did you produce the graph from a simulation or actual data?
I'd like to try that with my valves.......

[iamnothim]

Maybe something with a higher current rating considering the inrush (surge) current when the valve is cold.  4 Amp bridges are still cheap.

David,
So the one above won't work.  It's 50A.  I need 2 max.
I probably not looking at the right component.

[IanB]

Did you produce the graph from a simulation or actual data?

That was actual data. I used a variable voltage power supply and recorded the voltage and the current for each point.

I'd love to see it over time.

My graph shows steady values with time not involved.

It would be difficult for me to plot a graph showing the current surge at power on as the cold resistance is 0.98 ohms. If I connected a 12 V supply to the bulb the initial current would be over 12 A and I don't have a power supply readily to hand that is capable of that. I would have to make such a supply using a high power battery. However, given that it would be a fun experiment I might give it a try sometime.

[David_AVD]

Maybe something with a higher current rating considering the inrush (surge) current when the valve is cold.  4 Amp bridges are still cheap.

David,
So the one above won't work.  It's 50A.  I need 2 max.
I probably not looking at the right component.

That 50A figure is a peak rating (for one AC cycle) and not continuous.  Your valve surge will be brief, but still much longer than the time allowed for in the peak rating of that 1A bridge linked to.

[iamnothim]

I think I hear calculus.  We're not mutiplin' and divid'n anymore. Curves, acceleration, n stuff.

My only hope is electronic toolbox pro on my iPad.

[westfw]

As the DC out value for a full wave bridge rectifier is 0.9xVAC you get (for approx 7VAC) 6.3VDC out

Since when?  AC will typically be measured in RMS, which is 1/sqrt(2) times half the peak-to-peak voltage.  For the no-load case, rectified and filtered voltage is 1.4 * V_RMS, or nearly 9V for a 6.3V transformer.  (minus diode drops, though...)

Now, for something like heating a filament, the relevant piece of data should be the POWER.  And the way RMS voltage is defined, the (averaged over time) power you get from a 6.3V RMS AC voltage (~9V peak), is supposed to be the same as you'd get from a 6.3V DC voltage (constant 6.3V)

[IanB]

I guess with a full wave bridge there will be two diode drops of 0.5 to 1.0 V each, so that would drop 7 V AC down to about 6 V rectified DC. But of course there is no precision about this voltage drop, it is just depends on the choice of diodes.

[iamnothim]

Hello.
I thought the no-load tests weren't important.
In case they are needed for a rectifier bridge, I just took no load measurements of my Hammond mfg. transformer,  Model 166G6.   115V AC    3.78 VA   6.3V  C.T.  0.600 A.    I have another Hammond that is 6A

My Fluke 287 measures "True RMS" 
With the meter set to "Peak" and probed across unloaded secondaries it recorded a Peak Max of 10.67V   after that it was 7.87V
When set to "min/max" mode it recorded a max of 7.809

EDIT.  With a 6BN8 valve in the circuit.  6.3V  and 0.600A
Peak E across the load was 9.11V   Peak I was 1.49A  wow.
E settled in to 6.71V
I settled in to .649
Very good numbers wo added resistance.

[iamnothim]

Update: (should anyone care)

Thanks guys for pointing me in the right direction.  I think I'm close.  I actually read your posts this time. Went back several times as a reference.
David, your comment on surge current helped.  I get it.  I used the fluke in record mode and saw the 1.4A surge.  I lasted 2 seconds.
As you said….  One cycle surge…… math….. 2 sec. = more than 1 cycle.
Now I just have to find a rectifier that will do that. ( stocked, that's priced reasonably.)

I also calculated the smoothing capacitor and a regulator.
This site helped me get the concept and the math:   http://www.electronics-tutorials.ws/diode/diode_6.html

 I made the calculations from measurements using the Peak function of the Fluke 287.
According to Fluke the voltage measurements are "True RMS"   From what I have learned Peak is more useful than min/max. 
That's my biggest concern.   Am I using the right measurements for Vmax and Vrmns.

Even if it doesn't work I want to tinker with it.  (When I find the right component.)

[AlfBaz]

As the DC out value for a full wave bridge rectifier is 0.9xVAC you get (for approx 7VAC) 6.3VDC out

Since when?  AC will typically be measured in RMS, which is 1/sqrt(2) times half the peak-to-peak voltage.

At low voltages the diode's non-linearity is sufficient to maker sinusoidal RMS shortcut formula's inaccurate. I'll admit the formula I gave is a kludge at best but at these values is closer to the real RMS value than 1/sqrt2. At higher voltages the latter is almost spot on.

  For the no-load case, rectified and filtered voltage is 1.4 * V_RMS, or nearly 9V for a 6.3V transformer.  (minus diode drops, though...)

I don't think I mentioned anything about filtering

[iamnothim]

I ordered a single-phase Rectron rectifier that supports 3A
I also got the smoothing Cap and Linear Regulator so I have something to look at with my scope.